Beijing/w and major spoligotype families of Mycobacterium tuberculosis strains isolated
from tuberculosis patients in Eastern Turkey
Baris Otlu1, Riza Durmaz1, Selami Gunal1, Christophe Sola2, Thierry Zozio2, Nalin Rastogi2
1Department of Clinical Microbiology, Faculty of Medicine, Inonu University, Malatya, Turkey;
2Unité de la Tuberculose et des Mycobactéries, Institut Pasteur de Guadeloupe, Guadeloupe
Corresponding author Baris Otlu
Department of Clinical Microbiology Inonu University Faculty of Medicine 44069 Malatya, Turkey
E-mail: [email protected]
INTRODUCTION
Although it is a historical disease, we are still far from achieving success in the fight against tu- berculosis.
About one third of the overall population throughout the world is infected by M. tubercu- losis and each year 2 million individuals die from tuberculosis (WHO, 2005). Accurate diagnosis, treatment and epidemiology of tuberculosis are essential to control the disease. Development of novel molecular typing methods contributes to the classical knowledge on the epidemiology of
tuberculosis (Qian et al., 1999; Sola et al., 2001;
van Soolingen et al., 1995).
There are a number of methods for the molecu- lar typing of tuberculosis bacilli, such as IS6110 restriction fragment length polymorphism (RFLP), spoligotyping, pTBN12 fingerprinting, and mycobacterial interspersed repetitive unites- variable numbers of tandem repeat (MIRU-VN- RT) typing (Cavusoglu et al., 2006; Mokrousov et al., 2004; Sola et al., 2003). The spoligotyping method is based on displaying the heterogeneity caused by spacers and repetitive sequences in the polymorphic chromosomal DR (direct repeat) lo- cus (Kamerbeek et al., 1997). Researches have shown that spoligotyping is a practical and rap- id method in both clinical and molecular epi- demiology. Databanks including spoligotyping re- sults are formed with an aim to investigate the worldwide distribution of the detected clades and to compare them with each other. The SpolDB4, The aim of this study was to determine the Beijing/W family and major phylogenetic clades of Mycobacterium tu- berculosis strains of tuberculosis patients in a city with a tuberculosis incidence higher than the country average. A total of 220 M. tuberculosis strains isolated over a period of more than four years were typed by spoligotyping.
Spoligotyping resulted in 64 different patterns, 38 (17.3%) of which were unique, and 26 were clusters including 182 (82.7%) strains. The major shared types were ST 53 (n=55, 25%), ST 41 (LAM7-TUR; n=19, 8.6%), and ST 284 (n=15, 6.8%). The major clades observed ranked in the following order: ill-defined T superfamily (n=112, 50.9%); Latino- American-Mediterranean (LAM; n=33, 15%); Haarlem (n=24, 10.9%); and the S family (n=9, 4.1%). Three strains were in the Beijing family. A high number of strains (33 strains) showed patterns that did not fall within any of the ma- jor clades described. M. tuberculosis strains in Malatya have both STs showing a widespread distribution over the world and those restricted to this city, confirming the highly diverse nature of tuberculosis. Our results suggest that the Beijing clade, which is more prevalent among the strains with MDR and isoniazid resistance, is currently not a problem in Eastern Turkey.
KEY WORDS: Tuberculosis, Turkey, Molecular epidemiology, Spoligotyping, Beijing genotype SUMMARY
Received January 07, 2009 Accepted March 02, 2009
which was established to serve this goal, includes information on more than 45,000 isolates (Brudey et al., 2006).
Initial spoligotyping studies showed that one of the M. tuberculosis spoligotypes, Beijing, was pre- dominant in Beijing, China and other studies fol- lowed suggesting that it was also widely preva- lent throughout Asia and in the West Pacific (Caminero et al., 2001). Many reports from Germany, Italy, Russia, Estonia, South Africa and Columbia documented that the isolates identified as Beijing genotype were associated with multi- ple drug resistance (Caminero et al., 2001;
Drobniewski et al., 2002; Filliol et al., 2002; Filliol et al., 2003; Glynn et al., 2002; Lari et al., 2004;
Toungoussova et al., 2002; van Soolingen et al., 1995). Similarly, the multi-drug resistant “W”
genotype isolated from a tuberculosis epidemic affecting more than 350 individuals in New York was a member of the Beijing genotype clade (Bifani et al., 1996). Two additional studies from Vietnam found that the Beijing genotype was pre- dominant particularly among very young patients (Anh et al., 2000; Le et al., 2000). Overall, these reports suggest that the Beijing/W genotype has become widespread throughout the world, and that it has implications on drug resistance.
Studies on the molecular epidemiology of tuber- culosis in Turkey are primarily based on IS6110 RFLP and pTBN12 typing methods (Cavusoglu et al., 2006; Durmaz et al., 2003b; Durmaz et al., 2003a). Such reports contribute to our under- standing of the clonal relationship between M.
tuberculosis isolates in our city and throughout Turkey. However, our knowledge on the major
spoligotype profiles and particularly the Beijing/W genotype is limited. There is only one published study conducted on 147 M. tuberculo- sis strains collected from Malatya (Zozio et al., 2005). The study showed that three shared types (ST); ST41 (LAM7-TUR family), ST53 (ill-defined T1 superfamily), and ST50 (Haarlem 3 family) were the prevalent STs encountered in more than half the typed strains. The study described a new phylogeographically-specific clone called LAM7- TUR, which was a specific for Turkey, and sug- gested further studies be conducted on more strains to understand the real situation about the distribution of this STs in our city and country.
The aim of this study was to obtain information on the status of Beijing and other spoligotype clades among M. tuberculosis isolates collected within a period of more than four years in Malatya, a city in east Anatolia, which has a pop- ulation of 850,000, with 250-300 new tuberculo- sis cases each year, which exceeds the average tu- berculosis incidence of the country (32 vs. 26 per 100 000 persons, respectively).
MATERIAL AND METHODS
M. tuberculosis strains: A total of 220 strains iso- lated from patients diagnosed with M. tuberculo- sis primary infection attending Inonu University Turgut Ozal Medical Center, two tuberculosis dis- pensaries, and Governor Hospital in the period between January 2001 and July 2005 were in- cluded in the study. The total number of M. tu- berculosis strains isolated in the city throughout
TABLE 1 - Distribution of the 220 strains in according to study period and multi-drug resistance statue.
Year No of the strains No of the Sampling rate No of MDR
isolated/year strains analyzed (%) strains
2001 85 60 70.5 3
2002 126 52 41.3 4
2003 70 39 55.7 2
2004 95 55 57.9 4
2005 (first six months) 50 14 28.0 -
Total 426 220 51.6 13
this period was 426. Thus, 220 strains included in the study represented 51.6% of the overall iso- lates. All of the 220 patients were characterized by pulmonary TB and originated from Malatya (there were no foreign-born cases). The mean age was 33.2 years (range: 3-89), and the male-to-fe- male ratio was 1.4. The strains analyzed in this study included the isolates for which stock cul- tures were available and which were not analyzed in our published study (Zozio et al., 2005). Only one isolate of each patient was typed. The strains were identified based on their growth character- istics, biochemical tests such as nitrate reduction and niacin accumulation, and by the LCD-Array Chipron Myco-ID-Basic 1.2 (Chipron GmbH, Germany) or GenoType Mycobacterium CM (Hain Lifescience GmbH, Germany) kits, ac- cording to the manufacturer’s recommendations.
The distribution of strains according to years and the percentage of sampling are shown in Table 1.
Resistance tests for antimicrobial drugs The resistance of strains to isoniazid (INH), ri- fampicin (RIF), ethambutol (ETB) and strepto- mycin (SM) were evaluated using either Bactec 460 radiometric or Bactec MGIT 960 (Becton Dickinson and Company, Sparks, MD, USA) sys- tems, according to the manufacturer’s recom- mendations. Drug resistance was defined as greater than 1% growth in the presence of 0.1 mg of isoniazid per ml, 2 mg of rifampicin per ml, 2.5 mg of ethambutol per ml, or 2 mg of strepto- mycin per ml. Strains resistant to INH and RIF or additionally to one or more of the other drugs were considered multi-drug resistant (MDR) (Sharma and Mohan, 2004).
Spoligotyping
DNA was extracted from M. tuberculosis strains, grown in Lowenstein-Jensen medium, by the CTAB method (van Soolingen et al., 1991). M. tu- berculosis DNA was amplified using DRa (bi- otinylated 5’) and DRb primers as described by Kamerbeek et al., (Kamerbeek et al., 1997). The amplification product was hybridized with the membrane carrying probes (produced by Immunetics, Boston, USA) specific to the 43- spacer region and the product was visualized by radiation on x-ray film using chemiluminescence.
The data obtained were analyzed using the SpolDB4 (http://www.pasteur-guadeloupe.fr:8081/
SITVITDemo/outilsConsultation.jsp) database, and clusters of “shared-type (ST)” and clades were determined.
IS6110 RFLP typing IS6110-RFLP of PvuII-di- gested DNA was performed as previously de- scribed (Van Embden et al., 1993). After elec- trophoresis, the restriction fragments on the gel were denatured and blotted onto nylon mem- brane by the alkaline transfer procedure, and hy- bridization was carried out with a chemilumi- nescent 521 bp IS6110 fragment produced by PCR. M. tuberculosis H37Rv was used as an in- ternational standard. Banding patterns of the strains were analyzed with GelCompar II soft- ware (Version 3.5 Applied Math, Ghent, Belgium).
Statistical analysis
Molecular typing results were analyzed using Taxotron (Institut Pasteur, P. Grimont) and Bionumerics (v 3.1, Applied Maths, Sint Maarten Latem, Belgium). The pairwise distance between clinical isolates was computed using the 1- Jaccard index (Saitou and Nei 1987). The UPG- MA (unweighted pair-group method using arith- metic averages) was used for clustering of the iso- lates (Sneat and Sokal, 1973).
RESULTS AND DISCUSSION
Antibiotic resistance testing revealed that 13 out of 220 strains (5.9%) were multi-drug resistant (MDR), and 54 (24.5%) strains showed resistance to at least one drug. The rates of monoresistance to INH, RIF, ETB, and SM were 5.5% (12/220 strains), 2.7% (6/220 strains), 2.3% (5/220 strains), and 5.5% (12/220 strains), respectively. The re- maining 153 strains (69.5%) were susceptible to all drugs tested. Resistance testing has become critical for the control of tuberculosis due to in- creasing multi-drug resistance. Although many countries undertake resistance studies, the actu- al rate of resistance is still unknown. The rates of at least one drug resistance and multi-drug re- sistance in Turkey are 14.3-41% and 2.2-3% re- spectively (Durmaz et al., 2003b; Durmaz et al., 2003a).
Spoligotyping, which is rapid and convenient, en- sures the identification of ST and clades of strains (Sola et al., 2003; Yang et al., 1998). In the pres- ent study, typing of 220 strains resulted in 64 dif-
TABLE 2 - Orphan and newly-defined clusters in this study.
Strain Spoligotyping pattern Octal number Drug resistance STs No. of strain in this study
n (%)
N531 ■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■ 777767404760731 Susceptible N533 ■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■ 777767404760731 Susceptible
N534 ■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■ 777767404760731 Susceptible Unknown 6 2.7 (new cluster)
N547 ■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■ 777767404760731 Susceptible N566 ■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■ 777767404760731 Susceptible N571 ■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■ 777767404760731 Susceptible N433 ■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■ 777377777760601 Susceptible
N441 ■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■ 777377777760601 Non-MDRa Unknown 3 1.4 (new cluster)
N598 ■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■ 777377777760601 Susceptible
N410 ■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■ 777777637760771 Susceptible Orphan 1 0.5 N411 ■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■ 777777677760771 Susceptible Orphan 1 0.5 N426 ■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■ 777737777700171 Susceptible Orphan 1 0.5
N537 ■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■ 773760007760771 Non-MDRb Orphan 1 0.5
N543 ■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■ 377777763760771 Susceptible Orphan 1 0.5 N553 ■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■ 777737777700171 Susceptible Orphan 1 0.5 N561 ■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■ 037637477720771 Susceptible Orphan 1 0.5 N576 ■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■ 740617777760771 Susceptible Orphan 1 0.5 N579 ■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■ 777777777740171 Susceptible Orphan 1 0.5 N581 ■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■ 700377777760771 Susceptible Orphan 1 0.5 N595 ■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■ 377777763760771 Susceptible Orphan 1 0.5 N596 ■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■ 617777774020760 Susceptible Orphan 1 0.5 N606 ■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■ 773760007760771 Susceptible Orphan 1 0.5 N614 ■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■ 037637477720771 Susceptible Orphan 1 0.5 N964 ■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■ 377777763760771 Susceptible Orphan 1 0.5 N969 ■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■ 777771777760771 Susceptible Orphan 1 0.5 N971 ■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■ 716376777760771 Susceptible Orphan 1 0.5 N974 ■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■ 000000007740771 Non-MDRc Orphan 1 0.5 N170 ■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■ 037637477720771 Susceptible Orphan 1 0.5
N239 ■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■ 777000277740071 Non-MDRd Orphan 1 0.5
N241 ■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■ 037636377760771 Susceptible Orphan 1 0.5 N274 ■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■ 074002007760771 Susceptible Orphan 1 0.5 N291 ■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■ 377777763760771 Susceptible Orphan 1 0.5 N297 ■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■ 736377700000000 Susceptible Orphan 1 0.5
aResistance to isoniazid and streptomycin; bresistance to isoniazid; cresistance to rifampicin; dresistance to streptomycin.
TABLE 3 - Distribution of the 24 clusters with previously described STs.
Spoligotyping pattern Octal number ST Ratio in the 220 strains (%) Family
N %
■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■ 777777777760771 53 55 25.0 T1
■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■ 777777404760771 41 19 8.6 LAM7-TUR
■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■ 037637777760771 284 15 6.8 T1
■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■ 377777777760771 7 9 4.1 T
■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■ 777777737760771 86 8 3.6 T
■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■ 777777774020771 47 7 3.2 H1
■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■ 777777774020731 62 6 2.7 H1
■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■ 777777777720771 50 5 2.3 H3
■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■ 777717777760771 131 5 2.3 T3
■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■ 777777177760771 1252 5 2.3 X
■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■ 777777377760771 40 4 1.8 T
■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■ 776377777760731 784 4 1.8 S
■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■ 000000007760771 4 4 1.8 S
■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■ 000000004020771 2 3 1.4 H2
■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■ 777777404760760 1937 3 1.4 LAM7-TUR
■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■ 777777607760771 42 3 1.4 LAM9
■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■ 000000000003771 1 3 1.4 Beijing
■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■ 777777776360771 123 3 1.4 T1
■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■ 777777404760731 1261 2 0.9 LAM7-TUR
■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■ 577737777420771 361 2 0.9 H4
■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■ 177777777760771 191 2 0.9 T1
■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■ 777761007760771 766 2 0.9 LAM
■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■ 377777777760731 853 2 0.9 T2
■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■■ 777377777760771 1166 2 0.9 T1
Abbreviations, ST: Shared types, T: ill-defined T families, LAM: Latin-American-Mediterranean TUR: Turkey, H: Haarlem, X: X family, S: S family
ferent spoligotype patterns. While 182 strains were in the 26 clusters including 2 new clusters with 9 strains in this study, 38 (17.3%) strains dis- played unique profiles. Hence, the clustering rate of Malatya’s strains was 82.7%. Comparison of
the spoligotyping results obtained in our study with those in the SpolDB4 databank (http://www.pasteur-guadeloupe.fr:8081/SITVIT Demo) revealed that 24 of the 38 unique patterns (10.9 %) yielded actual orphan profiles that did
not cluster with any strain (Table 2). The re- maining 14 strains constructed a cluster with the strains in the SpolDB4 database.
The distribution of 24 clusters with known ST numbers is shown in Table 3. The most common STs were ST 53 (n=55, 25%), ST 41 (LAM7-TUR;
n=19, 8.6%), and ST 284 (n=15, 6.8%). The dis- tribution of spoligotyping profiles varies accord- ing to countries and regions. In our neighboring country Iran, the most common STs, ST 47, ST 11, ST 26, and ST 1, differed from the predomi- nant STs in Malatya, which is close to this coun- try (Farnia et al., 2006). In another study in Turkey, ST41, ST53 and ST50 genotypes were the most predominant strains in decreasing order in two different cities, Malatya and Ankara (Zozio et al., 2005). The predominant ST 53 strain in our study is also widespread throughout the world, particularly in European countries (Puustinen et
al., 2003). It was interesting to find that the ST 41 (LAM7-TUR), which is phylogeographically specific to Anatolia (Zozio et al., 2005), was the second most prevalent type in Malatya. The per- centage of ST 284, a globally rare (0.1%) type, was high (6.8%) in our city. Similar results were obtained in a previous study in Turkey with a 5.7% prevalence of ST 284 (Zozio et al., 2005). A few reports suggested that this genotype, which was also detected in Bulgaria and Saudi Arabia, may be phylogeographically specific to the Middle East (Zozio et al., 2005). To analyze the ST 41 (LAM7-TUR) and ST 284, which were the two prevalent types found in our study popula- tion, in detail IS6110 RFLP typing was per- formed. Of the 19 LAM7-TUR strains, 9 had few- er than 6 copies of IS6110 (low copy number strains), 10 had IS6110 copy numbers ranging from 6 to 13. Only one strain yielded a similar IS6110 RFLP profile with a strain identified as ST284, the remaining 18 strains had unique IS6110 RFLP profiles. Four of the 15 strains clas- sified as ST284 had low copy and 11 had 6-14 copies of IS6110. Only one strain having 6 copies had a similar IS6110 RFLP profile, the other strains were unique (Figure 1).
The most common clades in our study were ill-de- fined T clade (n=112, 50.9%), LAM7-TUR (n=26, 11.8%) (specific to our country), followed by Haarlem (n=24, 10.9%), S (n=9, 4.1%), X (n=5, 2.3%), LAM9 (n=3, 1.4%), and LAM (n=3, 1.4%).
The number of strains from the Beijing/W fami- ly was 3 (1.4%). In addition, one strain, each from CAS1- Delhi was detected. The clade names of 33 strains could not be identified (Table 4). The pre- dominant M. tuberculosis clades vary across pop- ulations. For example, about 50% of Beijing and Beijing-like strains are in the Far East, 25% of Haarlem clade is in Europe, and approximately 50% of LAM clade is in South America (Brudey et al., 2006). Haarlem, LAM and T clades were of- ten transiently detected in Africa, Central America, Europe and South America (Brudey et al., 2006; Zozio et al., 2005). T clade is detected in all continents and accounts for about 30% of strains in the databank. A study from Iran showed that the east African-Indian (24%), cen- tral Asian (20.8%), T clade (20.7%), Haarlem I (4.4%), and Beijing (3.2%) were the major M. tu- berculosis superfamilies (Velayati et al., 2006). The results of another study from Turkey (Gencer B FIGURE 1 - IS6110 RFLP profiles of the strains classi-
fied in ST41 (LAM7-TUR), ST284, and ST 1(Beijing).
H37Rv; Mycobacterium tuberculosis H37Rv strain used as a reference strain. Digits on top of the figure indicat- ed the strains numbers in each ST.
and Shinnick TM, 2005) indicating the predom- inance of T (37%), LAM (20%) and Haarlem (8%) clades, are in agreement with our findings.
However, while the Beijing clade ranked fourth in that study, we detected only three Beijing strains (1.4%). Two of the three Beijing strains had 10 copies of IS6110, the other had 9 copies of IS6110 and these strains represented unique RFLP profiles (Figure 1). Our results regarding the rate of Beijing genotype were similar to the study performed in which more than 4000 strains were collected from Istanbul/Turkey. This study showed that 1.1% of the strains were Beijing genotypes and the strains having Beijing geno- type were originated from the former Soviet Union (Koksalan et al., 2006). In reports from neighboring countries including different groups, the percentages of Beijing strains were 44.5% in Russia, 29.2% in Estonia, and 70.8% in Azerbaijan (Pfyffer et al., 2001; Sola et al., 2003).
Many studies from New York, Estonia, Cuba and Russia demonstrate a strong correlation between MDR and Beijing strains and indicate that this strain has a high virulence (Caminero et al., 2001;
van Soolingen et al., 1995). A study from China re- ported that the Beijing clade accounted for 77.8%
of multi-drug resistant strains (Le et al., 2000), another study from Azerbaijan detected a rate of 52.3% (Pfyffer et al., 2001). A study from Istanbul showed that Beijing genotype among at least ri- fampicin resistant strains was significantly more prevalent than among rifampicin-susceptible strains (Koksalan et al., 2006). However an asso- ciation with drug resistance could not be detect- ed for the Beijing strain in reports from Southeast Asia (Baker et al., 2005). Another important find- ing is that INH resistance alone was higher among Beijing/W and Beijing-like strains com- pared to others (Streicher et al., 2004). Of the three Beijing isolates in this study, two showed multi-drug resistance and one showed resistance to the combination of isoniazid and streptomycin.
When 13 strains with MDR were analyzed, two (15.4%) were of Beijing genotype. Although the number of Beijing strains is limited, our data are in agreement with the results of other reports sug- gesting an association of the Beijing genotype with MDR.
We detected only one Delhi isolate of the Central Asian (CAS) clade. In a study from Pakistan, the CAS clade was the predominant genotype with a rate of 39% (Hasan et al., 2006). Despite reports indicating that the CAS1-Delhi strain was strong- ly associated with isoniazid resistance (Baker et al., 2005), our strain was sensitive to all drugs.
The X clade is among the well-defined genotypes.
To date, three subclades, X1, X2 and X3 have been determined. X clade is the predominant genotype particularly in South America (14%) and United Kingdom (Brudey et al., 2006; Sola et al., 2003). We detected 5 strains (2.3%) from the X clade.
Similar to the results of the previous study in Turkey (Zozio et al., 2005), of the 220 strains an- alyzed, 33 (15%) did not correlate with any strain in the global databank and thus they had profiles specific to Malatya. This indicates that M. tuber- culosis strains in our region have high genetic heterogeneity. The philogenetic heterogeneity of tuberculosis, which is a major public health prob- lem in Turkey, may be attributed to the large pop- ulation (69.6 million according to the 2002 cen- sus) residing in a wide area (780,576 km2) ex- tending from Asia to Europe.
In conclusion, despite the predominance of TABLE 4 - Major spoligotyping families found
in this study.
Family No. of strains Percentage
in this study
T lineage 112 50.9
T 24 10.9
T1 81 36.8
T2 2 0.9
T3 5 2.3
LAM lineage 33 15.0
LAM 3 1.4
LAM7-TUR 26 11.8
LAM9 3 1.4
LAM5 1 0.5
Haarlem 24 10.9
S 9 4.1
X 5 2.3
Beijing 3 1.4
CAS1-Delhi 1 0.5
Unknown 33 15.0
spoligotypes such as Beijing originating from Asian countries, Malatya, which is close to the neighboring countries in Asia, has a very limited number strains identified as Beijing type and has a heterogeneous M. tuberculosis population com- prising globally distributed types and country- specific types.
ACKNOWLEDGEMENTS
This project was partly supported by TUBITAK (SBAG-HD-43 (105S152)), Ankara, Turkey. TZ was supported through a Ph.D. fellowship awarded by the European Union and the Regional Council of Guadeloupe and the International Network of the Pasteur Institutes.
REFERENCES
ANHD.D., BORGDORFFM.W., VANL.N., LANN.T., VAN
GORKOM, T., KREMERK.,VANSOOLINGEND. (2000).
Mycobacterium tuberculosis Beijing genotype emerging in Vietnam. Emerg. Infect. Dis. 6, 302- 305.
BAKERL.V., BROWNT.J., MAXWELLO., GIBSONA.L., FANG
Z., YATESM.D., DROBNIEWSKIF.A. (2005). Molecular analysis of isoniazid-resistant Mycobacterium tu- berculosis isolates from England and Wales reveals the phylogenetic significance of the ahpC -46A poly- morphism. Antimicrob. Agents Chemother. 49, 1455-1464.
BIFANIP.J., PLIKAYTISB.B., KAPURV., STOCKBAUERK., PANX., LUTFEYM.L., MOGHAZEHS.L., EISNERW., DANIELT.M., KAPLANM.H., CRAWFORDJ.T., MUSSER
J.M., KREISWIRTHB.N. (1996). Origin and interstate spread of a New York City multidrug-resistant Mycobacterium tuberculosis clone family. JAMA.
275, 452-457.
BRUDEYK., DRISCOLLJ.R., RIGOUTSL., PRODINGERW.M., GORIA., ALHAJOJ S.A., ALLIXC., ARISTIMUNOL., ARORA J., BAUMANIS V., ET AL. (2006).
Mycobacterium tuberculosis complex genetic di- versity: mining the fourth international spoligo- typing database (SpolDB4) for classification, pop- ulation genetics and epidemiology. BMC. Microbiol.
6, 23.
CAMINERO JA., PENA MJ., CAMPOS-HERRERO MI., RODRIGUEZ JC., GARCIA I., CABRERAP., LAFOZC., SAMPERS., TAKIFFH., AFONSOO., PAVONJM., TORRES
MJ., VANSOOLINGEND., ENARSONDA., MARTINC.
(2001). Epidemiological evidence of the spread of a Mycobacterium tuberculosis strain of the Beijing genotype on Gran Canaria Island. Am. J. Respir.
Crit. Care Med. 164, 1165-1170.
CAVUSOGLUC., TURHANA., AKINCIP., SOYLERI. (2006).
Evaluation of the Genotype MTBDR assay for rap- id detection of rifampin and isoniazid resistance in Mycobacterium tuberculosis isolates. J. Clin.
Microbiol. 44, 2338-2342.
DROBNIEWSKIF., BALABANOVAY., RUDDYM., WELDONL., JELTKOVAK., BROWNT., MALOMANOVAN., ELIZAROVA
E., MELENTYEYA., MUTOVKINE., ZHAKHAROVAS., FEDORINI. (2002). Rifampin- and multidrug-resist- ant tuberculosis in Russian civilians and prison in- mates: dominance of the beijing strain family.
Emerg. Infect. Dis. 8, 1320-1326.
DURMAZR., GUNALS., YANGZ., OZEROLI.H., CAVEM.D.
(2003a). Molecular epidemiology of tuberculosis in Turkey. Clin. Microbiol. Infect. 9, 873-877.
DURMAZR., OZEROLI.H., DURMAZB., GUNALS., SENOGLU
A., EVLIYAOGLUE. (2003b). Primary drug resistance and molecular epidemiology of Mycobacterium tu- berculosis isolates from patients in a population with high tuberculosis incidence in Turkey. Microb.
Drug. Resist. 9, 361-366.
FARNIAP., MASJEDIM.R., MIRSAEIDIM., MOHAMMADIF., JALLALEDING., VINCENTV., BAHADORI M., VELAYATI
A.A. (2006). Prevalence of Haarlem I and Beijing types of Mycobacterium tuberculosis strains in Iranian and Afghan MDR-TB patients. J. Infect. 53, 331-336.
FILLIOLI., DRISCOLLJ.R., VANSOOLINGEND., KREISWIRTH
B.N., KREMERK., VALETUDIEG., ANHD.D., BARLOW
R., BANERJEED., BIFANIP.J., ET AL. (2002). Global distribution of Mycobacterium tuberculosis spolig- otypes. Emerg. Infect. Dis. 8, 1347-1349.
FILLIOLI., DRISCOLLJ.R., VANSOOLINGEND., KREISWIRTH
B.N., KREMERK., VALETUDIEG., DANGD.A., BARLOW
R., BANERJEED., BIFANIP.J., ET AL. (2003). Snapshot of moving and expanding clones of Mycobacterium tuberculosis and their global distribution assessed by spoligotyping in an international study. J. Clin.
Microbiol. 41, 1963-1970.
GENCERB., SHINNICKT.M. (2005). Molecular epidemi- ology of Mycobacterium tuberculosis isolates from Turkey. Am. J. Infect. Dis. 1, 5-11.
GLYNNJ.R., WHITELEYJ., BIFANIP.J., KREMERK.,VAN
SOOLINGEND. (2002). Worldwide occurrence of Beijing/W strains of Mycobacterium tuberculosis:
a systematic review. Emerg. Infect Dis. 8, 843-849.
HASAN Z., TANVEER M., KANJI A., HASAN Q., GHEBREMICHAELS., HASANR. (2006). Spoligotyping of Mycobacterium tuberculosis isolates from Pakistan reveals predominance of Central Asian Strain 1 and Beijing isolates. J. Clin. Microbiol. 44, 1763-1768.
KAMERBEEKJ., SCHOULSL., KOLKA., VANAGTERVELDM.,
VAN SOOLINGEN D., KUIJPER S., BUNSCHOTEN A., MOLHUIZENH., SHAWR., GOYALM., VANEMBDENJ.
(1997). Simultaneous detection and strain differ- entiation of Mycobacterium tuberculosis for diag- nosis and epidemiology. J. Clin. Microbiol. 35, 907- 914.
KOKSALANO.K., KILICASLANZ., ZANLIERG., GUZELR., SEBERE. (2006). Prevalence of Beijing genotype Mycobacterium tuberculosis strains in Istanbul.
Int. J. Tuberc. Lung Dis. 10, 469-472.
LARIN., RINDIL., BONANNID., TORTOLIE., GARZELLIC.
(2004). Beijing/W Mycobacterium tuberculosis in Italy. Emerg. Infect Dis. 10, 958-959.
LET.K., BACHK.H., HOM.L., LEN.V., NGUYENT.N., CHEVRIERD., GUESDONJ.L. (2000). Molecular fin- gerprinting of Mycobacterium tuberculosis strains isolated in Vietnam using IS6110 as probe. Tuber.
Lung Dis. 80, 75-83.
MOKROUSOV I., NARVSKAYA O., LIMESCHENKO E., VYAZOVAYA A., OTTENT., VYSHNEVSKIY B. (2004).
Analysis of the allelic diversity of the mycobacter- ial interspersed repetitive units in Mycobacterium tuberculosis strains of the Beijing family: practical implications and evolutionary considerations. J.
Clin. Microbiol. 42, 2438-2444.
PFYFFERG.E., STRASSLEA., VANGORKUMT., PORTAELSF., RIGOUTSL., MATHIEUC., MIRZOYEVF., TRAOREH., VANEMBDEN J.D. (2001). Multidrug-resistant tu- berculosis in prison inmates, Azerbaijan. Emerg.
Infect. Dis. 7, 855-861.
PUUSTINENK., MARJAMAKI M., RASTOGI N., SOLA C., FILLIOLI., RUUTUP., HOLMSTROMP., VILJANENM.K., SOINI H. (2003). Characterization of Finnish Mycobacterium tuberculosis isolates by spoligo- typing. J. Clin. Microbiol. 41, 1525-1528.
QIAN L., VAN EMBDEN J.D., VAN DER ZANDEN A.G., WELTEVREDENE.F., DUANMUH., DOUGLASJ.T. (1999).
Retrospective analysis of the Beijing family of Mycobacterium tuberculosis in preserved lung tis- sues. J. Clin. Microbiol. 37, 471-474.
SAITOUN, NEIM. (1987). The neighbor-joining method:
a new method for reconstructing phylogenetic trees. Mol. Biol. Evol. 4, 406-425.
SHARMAS.K., MOHANA. (2004). Multidrug-resistant tu- berculosis. Indian J. Med. Res. 120, 354-376.
SNEATP., SOKALR. (1973). Numerical taxonomy: the principles and practicles of classification. WH Freeman and Co, San Francisco.
SOLAC., FILLIOLI., GUTIERREZM.C., MOKROUSOV I., VINCENTV., RASTOGIN. (2001). Spoligotype data- base of Mycobacterium tuberculosis: biogeographic distribution of shared types and epidemiologic and phylogenetic perspectives. Emerg. Infect. Dis. 7, 390-396.
SOLAC., FILLIOLI., LEGRANDE., LESJEANS., LOCHTC., SUPPLYP., RASTOGIN. (2003). Genotyping of the Mycobacterium tuberculosis complex using MIRUs: association with VNTR and spoligotyping
for molecular epidemiology and evolutionary ge- netics. Infect Genet. Evol. 3, 125-133.
STREICHERE.M., WARRENR.M., KEWLEYC., SIMPSONJ., RASTOGIN., SOLAC., VAN DERSPUYG.D., VANHELDEN
P.D., VICTORT.C. (2004). Genotypic and phenotyp- ic characterization of drug-resistant Mycobac - terium tuberculosis isolates from rural districts of the Western Cape Province of South Africa. J. Clin.
Microbiol. 42, 891-894.
TOUNGOUSSOVAO.S., SANDVENP., MARIANDYSHEVA.O., NIZOVTSEVAN.I., BJUNEG., CAUGANTD.A. (2002).
Spread of drug-resistant Mycobacterium tubercu- losis strains of the Beijing genotype in the Archangel Oblast, Russia. J. Clin. Microbiol. 40, 1930-1937.
VANEMBDENJ.D., CAVEM.D., CRAWFORDJ.T., DALEJ.W., EISENACHK.D., GICQUELB., HERMANSP., MARTINC., MCADAMR., SHINNICKT.M. (1993). Strain identifi- cation of Mycobacterium tuberculosis by DNA fin- gerprinting: recommendations for a standardized methodology. J. Clin. Microbiol. 31, 406-409.
VANSOOLINGEND., HERMANSP.W., DEHAASP.E., SOLL
D.R., VANEMBDENJ.D. (1991). Occurrence and sta- bility of insertion sequences in Mycobacterium tu- berculosis complex strains: evaluation of an inser- tion sequence-dependent DNA polymorphism as a tool in the epidemiology of tuberculosis. J. Clin.
Microbiol. 29, 2578-2586.
VANSOOLINGEND., QIANL., DEHAASP.E., DOUGLASJ.T., TRAOREH., PORTAELSF., QINGH.Z., ENKHSAIKAND., NYMADAWAP., VANEMBDENJ.D. (1995). Predomin - ance of a single genotype of Mycobacterium tu- berculosis in countries of east Asia. J. Clin.
Microbiol. 33, 3234-3238.
VELAYATI A.A., FARNIAP., MIRSAEIDIM., REZA M.M.
(2006). The most prevalent Mycobacterium tuber- culosis superfamilies among Iranian and Afghan TB cases. Scand. J. Infect. Dis. 38, 463-468.
WORLDHEALTHORGANIZATION(2005). WHO/HTM/TB.
Global tuberculosis control: Surveilance, Planning, Financing. WHO Report. Geneva.
YANGZ., BARNESP.F., CHAVESF., EISENACHK.D., WEIS
S.E., BATESJ.H., CAVEM.D. (1998). Diversity of DNA fingerprints of Mycobacterium tuberculosis isolates in the United States. J. Clin. Microbiol. 36, 1003- 1007.
ZOZIOT., ALLIXC., GUNALS., SARIBASZ., ALPA., DURMAZ
R., FAUVILLE-DUFAUXM., RASTOGIN., SOLAC. (2005).
Genotyping of Mycobacterium tuberculosis clini- cal isolates in two cities of Turkey: description of a new family of genotypes that is phylogeographi- cally specific for Asia Minor. BMC. Microbiol. 5, 44.
